Received: 30 October 2018 | Revised: 15 August 2019 | Accepted: 18 August 2019 DOI: 10.1002/ajp.23046 RESEARCH ARTICLE Extensive variability in the gut microbiome of a highly‐specialized and critically endangered lemur species across sites Mariah E. Donohue1,2 | Abigail E. Asangba3 | Jocelyn Ralainirina4 | David W. Weisrock2 | Rebecca M. Stumpf3 | Patricia C. Wright1,5 1Department of Ecology and Evolution, Stony Brook University, Stony Brook, New York Abstract 2Department of Biology, University of Deforestation continues to jeopardize Malagasy primates as viable habitats become Kentucky, Lexington, Kentucky smaller, more fragmented, and more disturbed. This deforestation can lead to 3Department of Anthropology, University of Illinois, Urbana, Illinois changes in diet, microhabitat, and gene flow between populations of endangered 4Department of Biological Anthropology, species, and it remains unclear how these changes may affect gut microbiome (GM) University of Antananarivo, Antananarivo, characteristics. The black‐and‐white ruffed lemur (Varecia variegata), which is among Madagascar ’ 5Department of Anthropology, Stony Brook Madagascar s most threatened lemur species, provides a critical model for under- University, Stony Brook, New York standing the relationships between historical and on‐going deforestation (habitat Correspondence disturbance), feeding ecology, and GM composition and diversity. We studied four Mariah E. Donohue, Department of Biology, populations inhabiting two rainforests (relatively pristine vs. highly disturbed) in 101 Morgan Building, University of Kentucky, ‐ Lexington 40506, KY. southeastern Madagascar. We conducted full day focal animal behavioral follows and Email: [email protected] collected fecal samples opportunistically across a three‐month period. Our results Funding information indicate that lemurs inhabiting sites characterized by habitat disturbance and low National Geographic Society, Grant/Award dietary diversity exhibited reduced gut microbial alpha diversity. We also show that Number: 9819‐15; Woese Institute for Genomic Biology; University of Illinois at these same factors were associated with high community dissimilarity using weighted Urbana Champaign; Directorate for Biological and unweighted UniFrac metrics. Finally, an indicator species analysis showed that Sciences, Grant/Award Numbers: 0820709, 0935347; Primate Conservation, Grant/ the most pristine site was characterized by an abundance of methanogenic archaea. Award Number: 74892 While it is impossible to disentangle the relative contributions of each confounding variable presented by our sampling design, these results provide crucial information about GM variability, thereby underscoring the importance of monitoring endangered species at the population‐level. KEYWORDS dietary diversity, gut microbiome, habitat disturbance, lemur 1 | INTRODUCTION changes in the relative abundance and diversity of these microbes can trigger deleterious effects on individual health and population Elucidating the factors shaping gut microbiome (GM) patterning in viability (Brucker & Bordenstein, 2012; Dethlefsen, Eckburg, Bik, & wild populations enhances our ability to conserve and monitor Relman, 2006; Flint, Scott, Louis, & Duncan, 2012; Hooper, Littman, & endangered species. Hosts outsource many digestive and immuno- Macpherson, 2012; Sekirov, Russell, Antunes, & Finlay, 2010). logical functions to symbiotic gastrointestinal microbes (e.g., mem- Therefore, understanding variation in GM responses to ecological bers of the GM; Amato et al., 2013; Bauchop, 1971; Flint, Bayer, factors is of particular interest for species facing ongoing anthro- Rincon, Lamed, & White, 2008; Hird, 2017; Lambert, 1998), and pogenic habitat alteration (Amato et al., 2016; Stumpf et al., 2016). Am J Primatol. 2019;e23046. wileyonlinelibrary.com/journal/ajp © 2019 Wiley Periodicals, Inc. | 1of12 https://doi.org/10.1002/ajp.23046 2of12 | DONOHUE ET AL. Several studies show that diet plays a crucial role in shaping the 2 | METHODS GM throughout an individual’s life (Clayton et al., 2016; David et al., 2014; De Filippo et al., 2010; Trosvik, Rueness, de Muinck, 2.1 | Ethics statement Moges, & Mekonnen, 2018; Turnbaugh, Bäckhed, Fulton, & Gordon, The methods used for noninvasive fecal collections of wild primates 2008) and over evolutionary timescales (Gomez et al., 2015; were reviewed and approved by the Stony Brook University IACUC Groussin et al., 2017; Ley et al., 2008; Springer et al., 2017; Trosvik committee (IACUC #2016–2254—USDA—NF). Field data collection et al., 2018; Yildirim et al., 2010). Anthropogenic disturbance is protocols were also approved by MNP (Madagascar National Parks), expected to impact food availability in tropical habitats through the body governing research in Madagascar’s protected areas. This changes to microclimates that drive variation in species composi- research adhered to the American Society of Primatologists tion (Abbas et al., 2011; Arrigo‐Nelson, 2006; Herrera, Wright, Principles for the Ethical Treatment of Non‐Human Primates. Lauterbur, Ratovonjanahary, & Taylor, 2011). Some primate species mayactuallybenefitfromhabitat disturbance, as plant species diversity often increases and can provide a new array of food items 2.2 | Study sites (e.g. Colobus sp.:Chapman,Struhsaker,Skorupa,Snaith,&Rothman, 2010; Dunham, 2017; Cercopithecus sp.: Kaplin & Moermond, 2000; This study was conducted in the two southernmost habitats within Microcebus sp. and Cheirogaleus sp.: Crowley, Blanco, Arrigo‐Nelson, the V. variegata range: Ranomafana National Park and Manombo & Irwin, 2013). However, the ability to incorporate novel foods Special Reserve (Figure 1). Within each area, we identified a gradient likely depends on habitat, severity of disturbance, and the species’ of habitat disturbance based on local interviews, Landsat images, and degree of specialty. Accordingly, negative consequences of habitat published papers. disturbance on diet have been described in Propithecus edwardsii Ranomafana National Park (RNP; −21.27 latitude and 47.33 (Arrigo‐Nelson, 2006; Matos, 2017), Rhinopithecus bieti (Huang longitude; 800‐1,200 m), established in 1991, is a 43,500‐hectare et al., 2017), Hylobates lar and Presbytis melalophos (Johns, 1986), expanse of continuous montane rainforest in Madagascar’s Ateles geoffroyi (Chaves, Stoner, & Arroyo‐Rodríguez, 2011), and southeastern province of Fianarantsoa (Wright et al., 2011). Alouatta palliata (Asensio, Cristobal‐Azkarate, Dias, Vea, & Rodrí- Within RNP, groups were followed in a pristine site (Mangevo guez‐Luna, 2007; Dunn, Cristóbal‐Azkarate, & Veà, 2010), all of or RNP‐P) and a lightly disturbed site (Vatoharanana or RNP‐LD). which show reduced dietary diversity and increased foraging effort RNP‐PandRNP‐LD are located within RNP’ssouthern in smaller, more degraded forests. block, separated by 8.34 km of continuous primary rainforest Black‐and‐white ruffed lemurs (Varecia variegata) are a critically (Holmes et al., 2013). endangered primate species endemic to the eastern rainforests of Disturbance categories withinRNPweredesignatedbasedon Madagascar. As dietary specialists consuming primarily fruit—a rare park history. RNP‐P is a seldom visited section of RNP, containing trait in lemurs (Wright, 1999)—V. variegata are considered especially 425 km2 of pristine primary rainforest virtually unaffected by sensitive to anthropogenic disturbance (Balko, 1998; Britt, 2000; human activities such as logging and hunting (Baden, Webster, and Herrera et al., 2011; Ratsimbazafy, 2002) and are often the first Kamilar (2016); Mancini (2016); Matos, 2017). RNP‐LD was species to succumb to local extinction (White, Overdorff, Balko, & selectively logged in 1986, during which time over 1,000 trees Wright, 1995). Deforestation has fragmented and degraded much of were removed from the 3.25 km2 area and an unspecified number their range, creating extreme population differentiation with very were damaged (Lehtonen, Mustonen, Ramiarinjanahary, Niemelä, limited gene flow in southern localities (Baden et al., 2014). & Rita, 2001). While RNP‐LD is typically categorized as “pristine” In this study, we aim to understand GM variability across distinct or “relatively pristine” primary forest, here it is designated as V. variegata populations. We selected groups occupying forests with “lightly disturbed” or “recovering” because it is among the most different histories and intensities of habitat disturbance to determine disturbed sections of V. variegata’s range within RNP (Herrera if different populations exhibit distinct GM characteristics, with the et al., 2011). In addition, many of the tree species that were ultimate goal of using this data to inform conservation efforts. removed are known V. variegata food resources, including However, we acknowledge that many environmental factors—both Canarium madagascariensis (ramy), Ocotea sp. (varongy), Crypto- related to and divorced from habitat disturbance—can influence the carya sp. (tavolo), and Chrysophyllum boivinianum (rahiaka; Arrigo‐ GM, and therefore simply aim to describe GM variation within and Nelson, 2006), highlighting the disproportionate effect of habitat between populations. We sampled from two relatively pristine sites disturbance on this highly specialized species. Since 1986, RNP‐LD within Ranomafana National Park’s continuous southern rainforest has received very little human disturbance. block, and two sites connected by a narrow corridor with high rates Manombo Special